Quantifying Lung Lesions in COVID-19

The data was collected from City of Hope patients during the period of 01 January 2020 till 30 June 2021. Volumetric quantification of lung and pneumonia lesions was carried out using the deep-learning segmentation model of lung and lesions on the 3DQI platform (https://3dqi.mgh.harvard.edu, accessed on 30 June 2021) [18 (link)]. After automated segmentation of lung and lesions, an image analyst, who was blinded to clinical data, reviewed and confirmed the segmentation results. A 3D radiometric analysis of the data was performed on the 3DQI Radiometrics tool based on extracted image features and clinical data.